I have a task to set a performance benchmark using stat given by google analytics.

Google Analytics stats: enter image description here

My Research:-

What I did was Since daily user count is 122154 I set benchmark as follows: I set user load as 5100 since (122154/24 = 5090)

enter image description here

Question:- Kindly help me to verify whether my approach is correct


Load is rarely the same during a day. So dividing a day load by 24 to get an hour load is not realistic. Try to get more detailed load statistics from Google Analytics. Some webapplications get 90% of a days load in one hour.

  • Thank you so much @Patrick. Is it practical to run like 50,000 user load using Jmeter?
    – Ted87
    Aug 30 '18 at 11:46

Your calculation might not be accurate as the load might not be evenly distributed during the day, for example if your application is targeting one country there could be "dead" periods like nights or weekends and there could be "active" periods when nearly all 120 000 users are online. So you need to narrow down monitoring telemetry time frame to 24 hours, identify how does load profile look like in terms of normal and peak loads and create 2 test types:

  1. Load test - simulate "normal" daily (or anticipated) load
  2. Stress Test - simulate peak load. You can also try to find the breaking point by increasing the load unless response time starts exceeding acceptable threshold or errors start occurring (whatever comes the first) to measure the maximum amount of users which your system can support.
  3. You can also consider Soak Testing - putting your system under a prolonged load to identify any potential memory leaks

Check out Why ‘Normal’ Load Testing Isn’t Enough article for more information

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.